Acetohydroxyacid synthase (AHAS; EC 4.1.3.18, also known as acetolactate synthase or ALS), is the first enzyme that catalyzes the biochemical synthesis of the branched-chain amino acids valine, leucine and isoleucine (Singh (1999) “Biosynthesis of valine, leucine and isoleucine,” in Plant Amino Acids, Singh, B. K., ed., Marcel Dekker Inc. New York, N.Y., pp. 227-247). AHAS is the site of action of four structurally diverse herbicide families including the sulfonylureas (LaRossa and Falco (1984) Trends Biotechnol. 2:158-161), the imidazolinones (Shaner et al. (1984) Plant Physiol. 76:545-546), the triazolopyrimidines (Subramanian and Gerwick (1989) “Inhibition of acetolactate synthase by triazolopyrimidines,” in Biocatalysis in Agricultural Biotechnology, Whitaker, J. R. and Sonnet, P. E. eds., ACS Symposium Series, American Chemical Society, Washington, D.C., pp. 277-288), and the pyrimidyloxybenzoates (Subramanian et al. (1990) Plant Physiol. 94: 239-244). Imidazolinone and sulfonylurea herbicides are widely used in modern agriculture due to their effectiveness at very low application rates and relative non-toxicity in animals. By inhibiting AHAS activity, these families of herbicides prevent further growth and development of susceptible plants including many weed species. Several examples of commercially available imidazolinone herbicides are PURSUIT® (imazethapyr), SCEPTER® (imazaquin) and ARSENAL® (imazapyr). Examples of sulfonylurea herbicides are chlorsulfuron, metsulfuron methyl, sulfometuron methyl, chlorimuron ethyl, thifensulfuron methyl, tribenuron methyl, bensulfuron methyl, nicosulfuron, ethametsulfuron methyl, rimsulfuron, triflusulfuron methyl, triasulfuron, primisulfuron methyl, cinosulfuron, amidosulfiuon, fluzasulfuron, imazosulfuron, pyrazosulfuron ethyl and halosulfuron.
Due to their high effectiveness and low-toxicity, imidazolinone herbicides are favored for application by spraying over the top of a wide area of vegetation. The ability to spray an herbicide over the top of a wide range of vegetation decreases the costs associated with plantation establishment and maintenance, and decreases the need for site preparation prior to use of such chemicals. Spraying over the top of a desired tolerant species also results in the ability to achieve maximum yield potential of the desired species due to the absence of competitive species. However, the ability to use such spray-over techniques is dependent upon the presence of imidazolinone-resistant species of the desired vegetation in the spray over area.
Among the major agricultural crops, some leguminous species such as soybean are naturally resistant to imidazolinone herbicides due to their ability to rapidly metabolize the herbicide compounds (Shaner and Robinson (1985) Weed Sci. 33:469-471). Other crops such as corn (Newhouse et al. (1992) Plant Physiol. 100:882886) and rice (Barrett et al. (1989) Crop Safeners for Herbicides, Academic Press, New York, pp. 195-220) are somewhat susceptible to imidazolinone herbicides. The differential sensitivity to the imidazolinone herbicides is dependent on the chemical nature of the particular herbicide and differential metabolism of the compound from a toxic to a non-toxic form in each plant (Shaner et al. (1984) Plant Physiol. 76:545-546; Brown et al., (1987) Pestic. Biochem. Physiol. 27:24-29). Other plant physiological differences such as absorption and translocation also play an important role in sensitivity (Shaner and Robinson (1985) Weed Sci. 33:469-471).
Plants resistant to imidazolinones, sulfonylureas and triazolopyrimidines have been successfully produced using seed, microspore, pollen, and callus mutagenesis in Zea mays, Arabidopsis thaliana, Brassica napus, Glycine max, and Nicotiana tabacum (Sebastian et al. (1989) Crop Sci. 29:1403-1408; Swanson et al., 1989 Theor. Appl. Genet. 78:525-530; Newhouse et al. (1991) Theor. Appl. Genet. 83:65-70; Sathasivan et al. (1991) Plant Physiol. 97:1044-1050; Mourand et al. (1993) J. Heredity 84:91-96). In all cases, a single, partially dominant nuclear gene conferred resistance. Four imidazolinone resistant wheat plants were also previously isolated following seed mutagenesis of Triticum aestivum L. cv. Fidel (Newhouse et al. (1992) Plant Physiol. 100:882-886) Inheritance studies confirmed that a single, partially dominant gene conferred resistance. Based on allelic studies, the authors concluded that the mutations in the four identified lines were located at the same locus. One of the Fidel cultivar resistance genes was designated FS-4 (Newhouse et al. (1992) Plant Physiol. 100:882-886).
Computer-based modeling of the three dimensional conformation of the AHAS-inhibitor complex predicts several amino acids in the proposed inhibitor binding pocket as sites where induced mutations would likely confer selective resistance to imidazolinones (Ott et al. (1996) J. Mol. Biol. 263:359-368). Wheat plants produced with some of these rationally designed mutations in the proposed binding sites of the AHAS enzyme have in fact exhibited specific resistance to a single class of herbicides (Ott et al. (1996) J. Mol. Biol. 263:359-368).
Plant resistance to imidazolinone herbicides has also been reported in a number of patents. U.S. Pat. Nos. 4,761,373, 5,331,107, 5,304,732, 6,211,438, 6,211,439 and 6,222,100 generally describe the use of an altered AHAS gene to elicit herbicide resistance in plants, and specifically discloses certain imidazolinone resistant corn lines. U.S. Pat. No. 5,013,659 discloses plants exhibiting herbicide resistance due to mutations in at least one amino acid in one or more conserved regions. The mutations described therein encode either cross-resistance for imidazolinones and sulfonylureas or sulfonylurea-specific resistance, but imidazolinone-specific resistance is not described. Additionally, U.S. Pat. No. 5,731,180 and U.S. Pat. No. 5,767,361 discuss an isolated gene having a single amino acid substitution in a wild-type monocot AHAS amino acid sequence that results in imidazolinone-specific resistance.
In plants, as in all other organisms examined, the AHAS enzyme is comprised of two subunits: a large subunit (catalytic role) and a small subunit (regulatory role) (Duggleby and Pang (2000) J. Biochem. Mol. Biol. 33:1-36). The large subunit (termed AHASL) may be encoded by a single gene as in the case of Arabidopsis and rice or by multiple gene family members as in maize, canola, and cotton. Specific, single-nucleotide substitutions in the large subunit confer upon the enzyme a degree of insensitivity to one or more classes of herbicides (Chang and Duggleby (1998) Biochem J. 333:765-777).
Mutations in genes encoding the AHAS large subunit, which are referred to herein as AHASL genes, are the molecular basis of herbicide tolerance in CLEARFIELD® crops which have increased tolerance to imidazolinone herbicides. Because each of these mutations results in a semi-dominant phenotype one mutation in a heterozygous state may be sufficient to produce a level of herbicide tolerance that is sufficient for many crop productions systems. However, for particular herbicide applications, and in cases with crop plants having multiple AHASL genes such as wheat, combinations of mutations are desired to achieve an increased level of resistance to herbicides.
For example, bread wheat, Triticum aestivum L., contains three homoeologous acetohydroxyacid synthase large subunit genes. Each of the genes exhibit significant expression based on herbicide response and biochemical data from mutants in each of the three genes (Ascenzi et al. (2003) International Society of Plant Molecular Biologists Congress, Barcelona, Spain, Ref. No. 510-17). The coding sequences of all three genes share extensive homology at the nucleotide level (WO 03/014357). Through sequencing the AHASL genes from several varieties of Triticum aestivum, the molecular basis of herbicide tolerance in most imidazolinone (IMI)-tolerant lines was found to be the mutation S653(At)N, indicating a serine to asparagine substitution at a position equivalent to the serine at amino acid 653 in Arabidopsis thaliana (WO 03/01436; WO 03/014357). The S653(At)N mutation in each gene is shaded in FIGS. 1A, 1B, and 1C. This mutation is due to a single nucleotide polymorphism (SNP) in the DNA sequence encoding the AHASL protein.
One goal of plant breeders is to introduce imidazolinone tolerance into existing wheat lines by inducing the S653(At)N mutation in the existing lines or by crossing non-IMI-tolerant lines with IMI-tolerant lines following by backcrossing and selection for imidazolinone tolerance. Another goal of plant breeders is to produce wheat plants with increased levels of imidazolinone tolerance, beyond the levels of tolerance seen in wheat plants possessing a single S653(At)N mutation in a single wheat AHASL gene. Thus, it is desirable to breed wheat plants that possess combinations of S653(At)N mutations at two or more of the AHASL genes. In addition, it is also desirable to breed wheat plants that are homozygous for the mutant S653(At)N allele at one or more of the AHASL genes. However, to develop the desired wheat plants, rapid methods for identifying the desired plants are needed. Existing methods of detecting wheat plants with imidazolinone tolerance are not well suited for use in the development of plants that possess more than a single S653(At)N allele at a single AHASL gene.
Existing methods of identifying plants with enhanced imidazolinone tolerance include field or greenhouse herbicide spray tests and biochemical assays for AHAS activity. Such methods are time consuming, however, and generally not suited for distinguishing, among large numbers of individual plants, subtle increases in imidazolinone tolerance that may occur when a second S653(At)N allele is introduced into a wheat plant.
Alternative methods for identifying desired plants include DNA-based methods. For example, the AHASL genes, or portions thereof, can be amplified from genomic DNA by polymerase chain reaction (PCR) methods and the resulting amplified AHASL gene or portion thereof can be sequenced to identify the mutant S653(At)N allele and the particular AHASL gene that it is present in. However, such a DNA-sequencing-based method is not practical for large numbers of samples. Another approach involves that use of radiolabelled or non-isotopically tagged, allele-specific oligonucleotides (ASOs) as probes for dot blots of genomic DNA or polymerase chain reaction (PCR) amplified DNA (Connor et al. (1983) Proc. Natl. Acad. Sci. USA 80:278-282; Orkin et al. (1983) J. Clin. Invest. 71:775-779; Brun et al. (1988) Nucl. Acids Res. 16:352; and Bugawan et al. (1988) Biotechnology 6:943-947. While such an approach is useful for distinguishing between two alleles at a single locus, this approach is not useful for the wheat AHASL genes, because three AHASL genes are nearly identical (FIG. 1) in region surround the SNP that gives rise to the mutant S653(At)N AHASL protein. Thus, a set of six oligonucleotide probes could not be developed that would be able to distinguish between the mutant and wild-type alleles at each of the three wheat AHASL genes.
One method that can be adapted for rapidly screening large numbers of individuals for the analysis of an SNP is the amplification refractory mutation system (ARMS) (Newton et al. (1989) Nucl. Acids Res. 17:2503-2516). This PCR-based method can be used to distinguish two alleles of a gene that differ by a single nucleotide and can also be used to distinguish heterozygotes from homozygotes for either allele by inspection of the PCR products after agarose gel electrophoresis and ethidium-bromide staining. The ARMS method is based on the premise that oligonucleotides with a mismatched 3′-residue will not function as primers in PCR under the appropriate conditions (Newton et al. (1989) Nucl. Acids Res. 17:2503-2516). While this method has proven useful for the analysis of an SNP at a single gene, whether this method, or a other PCR-based methods, can be used be used for the analysis of the SNP that gives rise to the S653(At)N mutation in each of the three wheat AHASL genes has not been reported.